We present a crustal deformation time-series and matching model for a full history of the 2018 Kilauea eruption using measurements from Sentinel-1 Interferometric Synthetic Aperture Radar (InSAR) and Global Positioning System (GPS). The eruption began on April 30, 2018 and continued for approximately 4 months, during which Kilauea’s summit crater and the East Rift Zone underwent continuous deflation and a Mw 6.9 earthquake struck the south flank on May 4. The InSAR data are processed with GMTSAR. We geometrically align all Sentinel-1 acquisitions using precise orbital information and select interferograms based on a 150-m perpendicular baseline and 60-day temporal baseline. Interferograms are masked based on stacked coherence, nearest-neighbor interpolated [Agram and Zebker, 2009] and then unwrapped with SNAPHU [Chen and Zebker, 2002]. We then use a remove/restore approach to force the line of sight displacement from each interferogram to match the point line-of-sight (LOS) measurements from 53 GPS time-series. This corrects some large spatial scale phase unwrapping errors due to atmospheric delay. Lastly, the time-series is constructed using coherence-based SBAS [Tong and Schmidt, 2016] to recover ground deformation. Results show that the summit crater of Kilauea and Pu’u’O’o both subsided more than a meter over 4 months (Figure).

We model the deformation using three types of sources. (1) The Mw6.9 earthquake is modeled as a low-angle thrust fault with a shallow landward dip and a depth of 2 km. The GPS and InSAR data spanning the May 4, 2018 event were used to develop the slip model and then it was removed from the displacement time-series. (2) Mass change measured by a repeated lidar survey of the subsidence at Halema‘uma‘u crater (https://www.opentopography.org) was used in a Boussinesq model to correct GPS and InSAR within a distance of 10 km of this crater. (3) After applying these two corrections, we model the remaining deformation using a large number of Mogi sources centered beneath Kilauea and along the East Rift Zone. Magma loss of the entire system is at least 0.3 km3, which is about one half the volume loss from the subsidence of the Halema‘uma‘u caldera of 0.61 km3. The total volume loss will be compared with the volume erupted along the lower southeast rift zone.

Several SAR sensors orbiting the Earth provide invaluable set of data since the nineties. Many applications benefit from the quasi monochromatic properties of the SAR signal to produce interferograms and compute deformation maps or Digital Elevation Models. SAR amplitude images are also used in diverse disciplines such as forestry, agriculture, archeology, geology, oceanography etc. for land cover changes detection, floods or wildfires mapping, oil spill detection, vessel tracking…

In the present contribution, we use the elementary property of SAR signal propagation to travel in straight line, independently from the time of the day or night, the weather conditions. Alike the visible light in optical images, the radar illumination of the ground in SAR images is hampered by sub vertical structure, and the shadow casted by the structure is proportional to the height of the structure. There is however a fundamental difference with optical images: within SAR images, points are located in the image depending on the signal travel time between the satellite and the target on the ground, while in the optical image it depends on the illumination angle. Nevertheless, knowing the incidence angle and the range resolution of the SAR sensor, one can deduce the height of the structure from the length of the shadow casted on the ground.

We apply that principle for monitoring of the lava lake level changes and the rise of the bottom of the crater at Nyiragongo volcano in in Democratic Republic of Congo. Nyiragongo is a stratovolcano hosting the largest lava lake on Earth. The city of Goma (800.000 inhabitants) is located at the foot of the volcano, in a region affected by recurrent conflicts, and where remote sensing is hence a convenient tool in assessing and monitoring volcanic hazards.

We use SAR images because the quasi permanent cloud cover blanketing this equatorial region and dense gas plume emanating from the crater hamper the use of optical or multispectral imagery.
From the frequent lava lake overflows, the bottom of the crater rose of more than 200 meters in a few years. The level of the lava lake is also affected by rapid level changes up to several tens of meters in a few hours or days. Despite the importance of these level changes and their evident relation with pressure changes in the magmatic system of the volcano, no permanent monitoring system was installed to measure these changes.

We measure the rise of the crater floor and the lava lake level fluctuations from the module of hundreds of SAR images acquired by RADARSAT, COSMO-SkyMed, SENTIEL-1 and Envisat. The height changes are obtained from the length of the shadow casted by surrounding rims. We set an automatic method for detecting illumination/shadow transitions using a grid-search strategy among an extensive dictionary of synthetics 1-D transects. We produced the first time series of accurate and dense lava lake level measures as well as of crater’s bottom altitude changes. The measures were validated with high resolution DEMs obtain from UAV photogrammetry and field measurements.

After its drainage in 2002, the crater filled up on ~400m thickness until 2008. Till 2012, the activity remained reduced (level fluctuations of various amplitude and time scales within the pit), then, after a large volume eruption at Nyamulagira volcano (located 13km away), the level progressively lowered, reaching ~70m below the crater floor. It remained at low level until the end of 2015. Early in 2016 a new vent opened within the crater, which intermittently emits important lava flows that blanket the crater. In November 2016, the lake level dropped of a hundred meters in a few hours.

We compare these measures with other parameters (seismicity, SO2 degassing, ground deformation and visual observations). We infer that these lava lake level changes can be used as a proxy for the pressure changes in the magmatic system. Comparing these pressure variation measurements with ground deformation measured by the permanent geodetic KivuGnet GNSS network, we present the first evidence of interaction between these two volcanoes, Nyiragongo and Nyamulagira, located 13 km apart.

The method of SAR shadow measurement presented here can of course be applied to several other type of height changes measurements such as sediment accumulation, water level variations in reservoirs etc. as long as there is a sub vertical structure casting shadow (e.g. cliff, rim, tower, dams etc.). This of course includes the monitoring of some of the other rare volcanoes hosting a lava lake, such as Kīlauea or Erta Ale.

The experimental small satellite Technology Experiment Carrier-1 (TET-1) is one of two satellites of the German Aerospace Center (DLR) FireBIRD mission. Its repetition rate is approximately five days at the equator. TET-1 operates two infrared cameras, one in the mid wave infrared and one in the long wave infrared, as well as a three-channel camera in the visible (red and green) and near infrared. Due to the higher spatial resolution and higher sensitivity with regard to thermal anomalies, with data from TET-1, a more detailed analysis of volcanic activity is possible than with data from other satellite missions such as the Moderate Resolution Imaging Spectroradiometer (MODIS).
Here, we present an atmospheric correction of the TET-1 data, based on the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) Global Emissivity Database (GDEM) and MODIS water vapour data with the shortest temporal baseline to the TET-1 acquisitions. Next, area coverage, the temperature, and radiant power of the detected thermal hotspots were derived at subpixel level and compared with observations derived from Visible Infrared Imaging Radiometer Suite (VIIRS) and MODIS data.
As a case study, a time series of nine TET-1 acquisitions before and after the 3 March 2015 eruption of the Villarrica Volcano was analysed. This Chilean volcano is one of the most active volcanoes in the South Andes Volcanic Zone. First thermal anomalies were detected nine days before the eruption. After the first eruption on 3 March 2015, a decrease of the radiant power was observed. This was followed by a stronger increase of the radiant power on 25 April 2015.
In addition to the analysis of the thermal signals due to volcanic activity, we also studied the eruption-related ash coverage at the eastern flank of the ice-capped Villarrica Volcano. TET-1 images acquired before and after the 3 March 2015 eruption were compared to map this ash coverage. Higher resolution optical data from Landsat-8 were used to validate the results obtained from TET-1 imagery.
Strengthened by the second satellite, the Berlin InfraRed Optical System (BIROS), which was launched in June 2016, the FireBIRD mission will in future support and complement ground-based observations of active volcanoes.

Satellite data are commonly used to support international disaster risk management efforts, such as those deployed to respond to the Indonesia, Sulawesi 2018 earthquake, and for the understanding of the geophysical phenomena behind the seismic event. The current work present a collaborative effort within the frame of the Geohazards Office, an initiative led by the French Geological Survey and supported by the European Space Agency (ESA) and the French Space Agency (CNES), for the scientific animation of EO practitioners and the exploitation of Earth Observation processing services to support hazard mapping and risk assessment.
Sulawesi lies within a complex fault system located between the Australian, Pacific, Philippine and Sunda Plates. The main active structure onshore at the western part of Central Sulawesi is the left-lateral NNW-SSE trending Palu-Koro strike-slip fault that forms the boundary between the North Sula and Makassar blocks. On 28 September 2018, a large tsunamigenic earthquake (Mw 7.5) struck the Minahasa Peninsula, Indonesia. The earthquake caused massive damages near Palu city, including onshore gravitational instabilities and a tsunami.
In order for us to characterize the seismogenic fault, we applied subpixel image correlation on Copernicus Sentinel-1 (06 July 2018 and 05 October 2018) and Sentinel-2 data (17 September 2018 and 2 October 2018) to derive the two-dimensional horizontal co-seismic displacement field. Our results, coupled with an elastic dislocation model, show a dominant senextral strike-slip motion on the onshore part of the Palu-Koro fault. Maximum displacement at the surface reached more than 8 meters at the location of Palu city. Our model suggests that significant deformation might also be absorbed by the underwater branch of the Palu-Koro fault; this evidence has implications regarding the mechanism of the associated tsunami.

A public national ground motion service, based on Copernicus Sentinel-1 data, was launched in Norway in November 2018. The service provides government, industry, scientists, and the general public with a consistent free and open InSAR based ground motion data source. The data are publicly available in a web browser interface, with tools for simple data analysis, as well as download.

The ground motion service has been built as part of Norway’s collaborative ground segment. The main motivation was to aid in the management of natural hazards, such as subsidence and landslides, as well as to help monitor critical infrastructure. In a country like Norway, with over 300,000 km2 of sparsely populated, mountainous landscape, earth observation obviously plays a crucial role. Ground motion data is critical throughout the process of natural hazards mapping and monitoring. In case of Norway, two prominent hazards addressed by this service are unstable slopes, and urban subsidence.

Motion maps are used to identify moving slopes which are then assessed to determine the main process driving the movement. During the hazard and risk classification stage, movement rates are fed into a semi-quantitative assessment process. Finally, medium and high risk sites are monitored over the long term, using both in situ instrumentation and InSAR data. In particular, for snow-covered areas, all high risk unstable mountain slopes are equipped with snow protected corner reflector (CR) networks.

While landslides represent the largest potential risk to lives, subsidence may be the natural hazard with the greatest economic impact, due to its impact on critical infrastructure. Rising sea levels threaten coastal cities around the world. In Norway, postglacial isostatic rebound is ongoing, and counteracts the effects of rising oceans. However, many of the coastal cities have expanded, building out into the sea on anthropogenic fill. In most of these areas, subsidence rates far exceed the rate of isostatic rebound, resulting in an increased threat from sea level rise. The data provided by the national ground motion service has the potential to be a very useful part of operational workflows for private companies in the construction business.

Natural hazards from extreme weather events, earthquakes, and volcanic eruptions have been on the rise in recent years, increasing the need for efficient solutions for the monitoring and management of triggered disasters. While Synthetic Aperture Radar (SAR) sensors, with their all-weather and day-and-night capabilities, have long been known as a useful data set in hazard monitoring, only the recently launched Sentinel-1 C-band SAR constellation has finally elevated SAR to a relevant data source for operational response.

In the work described in this paper, we utilize the emergency response capabilities of Sentinel-1 to develop an automatic hazard monitoring service that provides near real-time information from SAR for hazard events related to severe weather, earthquakes, and volcanic unrest. The presented work was accomplished in a close collaboration between the University of Alaska Fairbanks (UAF), the Alaska Satellite Facility (ASF), the NASA Short-term Prediction Research and Transition Center (SPoRT), the Alaska Volcano Observatory (AVO), and V-ADAPT, Inc., a company providing information on volcanic hazards to the aviation industry. The UAF team was responsible for the development of value-added products from SAR, targeting weather-related (hurricanes, tornadoes, flooding) and geophysical (earthquakes, volcanoes) hazards. UAF and ASF collaborated on the integration of these products into operational cloud-based processing pipelines, enabling automatic and near real-time data production over a target area. AVO and V-ADAPT assisted in the definition of value-added products for the volcano hazard community and included final products into their existing decision-support systems. NASA SPoRT, an intermediary organization specializing in transitioning remote sensing resources into operations, worked with end-user partners from the weather-related hazard community (e.g., the Federal Emergency Management Agency (FEMA)) on the integration of SAR data into FEMA disaster response activities.

The paper introduces the SAR-based value-added products (geocoded and radiometrically terrain corrected SAR image time series; InSAR-based deformation maps; change detection products; RGB composites from dual-pol Sentinel-1 information) that were developed for this effort. We describe the cloud-based production pipeline that was built within the Amazon Web Services (AWS) cloud environment to automatically produce these products in near-real time over hazard-affected regions. We describe our approach for integrating SAR-based data layers into end-user decision-making workflows and showcase the impact of SAR-derived data on the response to a number of recent hazard events. These events include the 2016 eruption at Pavlof, Alaska, the 2016/17 eruption at Bogoslof, Alaska, the 2018 Alaska North Slope earthquake, a 2017 Missouri/Mississippi river flood, the 2017 hurricanes Harvey, Irma, and Maria, the 2018 hurricanes Florence and Michael, as well as several 2017 tornado events in Mississippi and Wisconsin. Finally, we introduce qualitative and quantitative measures that are used to evaluate the performance of SAR-derived information layers in response situations. These measures assess how the products are helping end-user partners achieve their missions in a more efficient way, and quantify limitations of the system and products as identified by our partners.